Directory UMM :Data Elmu:jurnal:T:Tree Physiology:Vol15.1995:
Tree Physiology 15, 629--638
© 1995 Heron Publishing----Victoria, Canada
Physiological responses to water stress and waterlogging in Nothofagus
species
OSBERT J. SUN,1,2 GEOFFREY B. SWEET,1 DAVID WHITEHEAD3 and GRAEME D.
BUCHAN4
1
2
3
4
School of Forestry, University of Canterbury, Private Bag 4800, Christchurch, New Zealand
Present address: New Zealand Forest Research Institute, Private Bag 4800, Christchurch, New Zealand
Manaaki Whenua--Landcare Research, P.O. Box 69, Lincoln, Canterbury, New Zealand
Soil Science Department, Lincoln University, Canterbury, New Zealand
Received December 13, 1994
Summary Gas exchange and water relations were investigated in Nothofagus solandri var. cliffortioides (Hook. f.) Poole
(mountain beech) and Nothofagus menziesii (Hook. f.) Oerst
(silver beech) seedlings in response to water stress and waterlogging. At soil matric potentials (Ψsoil ) above − 0.005 MPa,
N. solandri had significantly higher photosynthetic rates (A),
and stomatal and residual conductances (gsw and grc), and lower
predawn xylem water potentials (Ψpredawn) than N. menziesii.
The relative tolerance of plants to water stress was defined in
terms of critical soil matric potential (Ψcri ) and lethal xylem
water potential (Ψlethal). The estimated values of Ψcri and Ψlethal
were −1.2 and − 7 MPa, respectively, for N. solandri, and −0.7
and −4 MPa, respectively, for N. menziesii. Photosynthesis was
sustained to a xylem water potential (Ψxylem) of − 7 MPa in
N. solandri compared with − 4 MPa in N. menziesii.
Following rewatering, both A and Ψxylem recovered quickly
in N. solandri, whereas the two variables recovered more
slowly in N. menziesii. During the development of water stress,
nonstomatal inhibition significantly affected A in both N. solandri and N. menziesii. Nothofagus menziesii was more susceptible to inhibition of A by waterlogging than N. solandri.
However, the tolerance of N. solandri to severe waterlogging
was also limited as a result of a failure to form adventitious
roots, suggesting a lack of adaptation to these conditions. The
differences in tolerance to water stress and waterlogging between the two species are consistent with the distribution
patterns of N. solandri and N. menziesii in New Zealand.
Keywords: gas exchange, geographical distribution, nonstomatal inhibition, Nothofagus menziesii, Nothofagus solandri
var. cliffortioides, water relations, water stress.
Introduction
Woody plants vary markedly in their response to and tolerance
of water stress (e.g., Myers and Landsberg 1989, Ranney et al.
1990, Seiler and Cazell 1990, Ni and Pallardy 1991) and
waterlogging (e.g., Pereira and Kozlowski 1977, Sena Gomes
and Kozlowski 1980, Tang and Kozlowski 1982, Pezeshki and
Chambers 1985). Water stress influences metabolism, physiology and morphology in plants (Passioura 1982). Waterlogging
causes inadequate aeration in soil, leading to rapid depletion
of oxygen which induces many physiological and morphological changes (Kozlowski 1984), and affects mineralization and
solubility of mineral substances, and leads to the formation of
phytotoxic compounds (Janiesch 1991).
Nothofagus solandri var. cliffortioides (Hook. f.) Poole and
Nothofagus menziesii (Hook. f.) Oerst are two of the five taxa
in the genus Nothofagus (Southern beech) that are native to
New Zealand. They both have a wide ecological and geographical distribution, occurring from sea level to the upper
timberlines at comparable latitudes. Taxonomically, however,
the two species belong to two subgroups with distinct pollen
types (Hanks and Fairbrothers 1976, Philipson and Philipson
1988, Hill and Read 1991). The genetic divergence of the two
species may be dated back to the upper Cretaceous (Wardle
1984), approximately 70 million years ago, before New Zealand was separated from Australia and Antarctica (Mildenhall
1980, McGlone 1985). From the present distribution, it appears that soil water availability and drainage may play a
prominent role in differentiating the geographical occurrence
of N. solandri and N. menziesii (Wardle 1984). Drought has
been suggested as an important factor in forest mortality that
has led to the present vegetation pattern in New Zealand (Jane
and Green 1986, Innes and Kelly 1992). Susceptibility of trees
to waterlogging followed by drought has also been linked to
forest decline (Jane and Green 1986). Plants that have been
waterlogged may be more sensitive to drought than plants
growing in aerated soils as a consequence of reductions in size
of the root systems in response to inundated soil conditions
(Jane and Green 1985, 1986). Nothofagus solandri is usually
more dominant than N. menziesii in habitats with a dry climate
(Wardle 1984), whereas on waterlogged soils, N. menziesii is
stunted and often replaced by species such as N. solandri and
Podocarpus dacrydioides A. Rich. (Wardle 1967). Despite
many studies of the ecology and ecosystem dynamics of Nothofagus-dominated forests in New Zealand, a detailed knowledge of water relations of Nothofagus under conditions of
630
SUN, SWEET, WHITEHEAD AND BUCHAN
water stress and waterlogging is lacking. To investigate the role
of water relations in determining the geographical distribution
of N. solandri and N. menziesii in New Zealand, we tested (1)
the responses of photosynthesis and xylem water potential to
soil water availability, (2) tolerance of seedlings to progressively increased soil drying, and (3) tolerance to waterlogging
in the two species.
Materials and methods
Response to water stress
Plant material, experimental design and treatment In the
spring of the first growing season, seedlings were raised from
seed and grown in pots containing a mix of peat (60%), soils
collected from a Nothofagus-dominated forest (20%), ground
pine bark (10%), coarse sand (10%) and slow-release fertilizer
(2 kg m −3, Osmocote® Plus, Grace-Sierra, Heerlen, The Netherlands). On March 20 (autumn) of the second growing season,
15 seedlings of each species were each transplanted to a 4-liter
pot containing nursery top soil (loam-textured) packed at a bulk
density (ρb) of approximately 1100 kg m −3. Leaf area (onesided) was estimated as 0.06 m2 for N. solandri and 0.05 m2 for
N. menziesii seedlings. The seedlings were arranged on a bench
in a greenhouse, and a dilute complete nutrient solution
(Ingestad 1971) containing 10 ppm nitrogen was applied regularly until treatments began. The photoperiod in the greenhouse
was extended to 16 h with 400 W sodium vapor lamps. Maximum photosynthetic photon flux density (Q) varied between
1000 µmol m −2 s −1 in midsummer and 600 µmol m −2 s −1 in
winter. Day/night temperatures (25/15 °C) were controlled
thermostatically.
Treatments began 4 weeks after transplanting. The experiment comprised three water stress treatments in a completely
randomized single-tree plot design, and each treatment contained five seedlings of each species. Seedlings in Treatment 1
were well watered throughout the experiment (control), to
maintain the soil matric potential (Ψsoil ) above −0.005 MPa.
Seedlings in Treatment 2 were rewatered on Day 30 when Ψsoil
reached a critical level (see below). Seedlings in Treatment 3
Figure 1. Relationship between matric potential (Ψsoil ) and volumetric
water content (θv) of the soil.
were left unwatered for 52 days until the end of the experiment.
All pots were placed randomly on the bench and rearranged
every second day.
Determination of soil matric potential Soil matric potential
(Ψsoil ) was estimated from volumetric soil water content (θv)
based on an equation describing the soil water characteristic
(Figure 1) determined using a tension table and pressure plate
apparatus (Buchan and Grewal 1990). Measurements of θv
were made with a time domain reflectometry (TDR) soil moisture meter (SoilMoisture Equipment Corp., Santa Barbara, CA,
USA).
Assessment of critical soil matric and plant water potentials
Tolerance to water stress was defined in terms of critical soil
matric potential (Ψcri ) and lethal xylem water potential (Ψlethal).
The value of Ψsoil at which Ψpredawn equals Ψmidday (of the
preceding day) is defined as Ψcri , i.e., the minimum Ψsoil for
adequate water uptake. When a plant reaches Ψcri , the hydraulic conductivity from the soil to leaves decreases abruptly
(Tyree et al. 1992), so that continuing dehydration causes Ψxylem
to decrease until leaf desiccation occurs. Survival and recovery
of the plant following rewatering may then depend on the
plant’s ability to tolerate Ψxylem that induces leaf desiccation.
The values of Ψxylem below which leaf desiccation occurred
were defined as Ψlethal.
Values of Ψcri for the two species were determined graphically from measurements of Ψpredawn and Ψmidday, and values of
Ψlethal were taken as the minimum Ψxylem before leaf desiccation became visible.
Measurements of photosynthesis and leaf conductance
Stomatal conductance to diffusion of water vapor (gsw), rates
of net photosynthesis (A) and intercellular CO2 concentration
(Ci) were measured at ambient CO2 concentration (about 380
µmol mol −1) in a controlled environment cabinet (Temperzone,
Auckland, New Zealand) between 1100 and 1400 h (NZ ST)
with a portable photosynthesis system (LI-6200, Li-Cor Inc.,
Lincoln, NE, USA). Measurements were made on twigs with
5--10 fully expanded leaves.
Seedlings were transferred from the greenhouse to the controlled environment cabinet at least 2 h before measurements
were made. Photosynthetic photon flux density (Q) in the
cabinet was 650 µmol m −2 s −1, and temperature and relative
humidity were maintained at 20 °C and 55%, respectively.
After each measurement, twigs were detached and immediately used for Ψmidday measurements. Leaf areas (one-sided)
were measured with an image area meter (Delta-T Devices,
Burwell, Cambridge, U.K.).
Leaf conductance was partitioned into stomatal (gsc) and
nonstomatal or residual conductance to diffusion of CO2 (grc)
according to Farquhar and Sharkey (1982), where the term grc
includes both mesophyll conductance (gmc) and carboxylation
efficiency (k).
Measurements of xylem water potential Measurements of
Ψxylem were made with a pressure chamber (PMS Instrument
Co., Corvallis, OR, USA). Measurements of Ψmidday were made
WATER STRESS AND WATERLOGGING IN NOTHOFAGUS
immediately after gas exchange measurements in the controlled environment cabinet between 1100 and 1400 h, and Ψpredawn
was measured in the greenhouse on the following day, about
14 h later. Twigs for measuring Ψxylem were obtained from a
second- or third-order branch in the middle of the main stem.
Data analysis Estimates of A, gsw, grc, Ψmidday and Ψpredawn in
Treatment 1 were made over a 52-day period. Results were
pooled for each species and compared by the Student’s t-test.
Values of Ψcri and Ψlethal were well separated between species
and were not compared statistically. Data for Treatments 1 and
3, and data from the early part of Treatment 2 (before soils were
rewatered) were used to determine the relationship between
Ψpredawn and Ψsoil , responses of A, gsw and grc to Ψpredawn, and
stomatal and nonstomatal (residual) components of photosynthesis. The data were evaluated by regression analysis and
analysis of variance.
631
test with a confidence level of P ≤ 0.05. Conditions in the
greenhouse on the measurement days are given in Table 2.
Results
Response to water stress
When Ψsoil was maintained above −0.005 MPa, A, gsw and grc
were significantly (P ≤ 0.001) greater by 30, 25 and 30%,
respectively, and Ψpredawn significantly lower by 30% for N. solandri than for N. menziesii (Table 3). However, the values for
Ψmidday did not differ between the species.
The effect of decreasing Ψsoil on Ψxylem differed according
to species (Figure 2). In N. solandri, decreases in both Ψmidday
and Ψpredawn occurred gradually in response to declining Ψsoil ,
whereas in N. menziesii, Ψmidday and Ψpredawn initially decreased slowly with decreasing Ψsoil but declined abruptly
when Ψsoil fell below −0.5 MPa. Values of Ψcri were estimated
as −1.2 and −0.7 MPa, and values of Ψlethal were estimated as
− 7 and − 4 MPa for N. solandri and N. menziesii, respectively.
In both species, values of A, gsw and grc declined with
decreasing Ψpredawn (Figure 3) following a relationship of the
form:
Response to waterlogging
Fifteen seedlings of each species were grown in pots containing potting mix as described previously. Soil water content of
the control group was maintained near field capacity throughout the experiment. The water table in the second group of
seedlings was raised 50 to 60 mm by placing pots in saucers
containing tap water (partial waterlogging). Seedlings in the
third group were totally flooded by submerging the pots individually in water containers (severe waterlogging). The water
in the containers was replaced daily with fresh tap water. All
pots were placed randomly on a bench in the greenhouse and
rearranged weekly. The air-filled porosity (εa) for the partial
waterlogging treatment was estimated to be only half of that of
the control (Table 1).
Sequential measurements of A and gsw were conducted on
twigs at ambient CO2 concentration between 1200 and 1400 h
in the greenhouse, and effects were evaluated by analysis of
variance. Means were compared by Duncan’s multiple-range
ln (y) = a − b ln|Ψ predawn |,
(1)
where b defines the sensitivity of the response. Values for
parameters a and b are summarized in Table 4. Both A and grc
were more sensitive to decreasing Ψpredawn (P ≤ 0.001) in
N. solandri than in N. menziesii, but there was no significant
difference for gsw.
Rapid reductions in A, gsw and grc occurred as Ψpredawn fell
from −0.4 to −0.8 MPa in N. solandri, and from −0.2 to −1.0
MPa in N. menziesii. In N. solandri, the reductions in A, gsw
and grc continued until Ψpredawn fell to − 7 MPa, whereas in
N. menziesii, A, gsw and grc initially declined slowly in response
Table 1. Soil bulk density (ρb), total porosity (ε), volumetric soil water content (θv), and air-filled porosity (εa) for three waterlogging treatments.
The values shown are means ± standard errors, n = 6.
Treatment
ρb (kg m − 3)
ε (%)
θv (%)
εa (%)
Control
Partial waterlogging
Severe waterlogging
776 ± 11
70.7 ± 0.04
50.2 ± 1.9
59.0 ± 0.3
70.7
20.5 ± 0.5
11.7 ± 0.7
0
Table 2. Mean photosynthetic photon flux density (Q), mean air temperature (Tair ), mean ambient CO2 concentration (Ca), and mean vapor pressure
deficit (D) in the greenhouse during the experimental period.
Days since start
of treatment
Q
(µmol m −2 s −1)
Tair
(°C)
Ca
(µmol mol − 1)
D
(kPa)
1
8
15
22
40
352
840
930
840
914
24.0
26.8
25.8
26.1
26.7
341
340
344
335
348
1.66
1.98
2.38
2.05
1.84
632
SUN, SWEET, WHITEHEAD AND BUCHAN
Table 3. Net photosynthetic rates (A), stomatal conductance to diffusion of water vapor (gsw), residual conductance to diffusion of CO2
(grc), and midday and predawn xylem water potentials (Ψmidday and
Ψpredawn) of seedlings at Ψsoil ≥ −0.005 MPa. The values shown are
means ± standard errors.
Variable
N. solandri
N. menziesii
A (µmol m −2 s −1)
gsw (mmol m −2 s −1)
grc (mmol m − 2 s − 1)
Ψmidday (MPa)
Ψpredawn(MPa)
10.61 ± 0.34*1
123.8 ± 4.0*
46.5 ± 2.0*
−1.15 ± 0.02
−0.52 ± 0.01*
8.17 ± 0.36
99.1 ± 5.2
34.5 ± 1.7
−1.14 ± 0.03
−0.35 ± 0.02
1
An asterisk indicates that the difference between species is significant at P ≤ 0.001.
species, b for the response of A to grc was more than 60%
greater than for the response of A to gsc.
When seedlings were rewatered after being subjected to
severe water stress (Ψcri ≤ −1.2 and −0.7 MPa for N. solandri
and N. menziesii, respectively), Ψxylem, A and gsw increased
more rapidly for N. solandri seedlings than for N. menziesii
seedlings (Figure 5). Three days after rewatering, both Ψmidday
and Ψpredawn had recovered by more than 2 MPa in N. solandri,
and Ψpredawn recovered to the control values after 14 days.
However, in rewatered seedlings that had been water stressed
previously, Ψmidday remained consistently lower than in control
seedlings throughout the experiment. In N. menziesii, the recovery of Ψxylem following rewatering was slower than in
N. solandri, especially for Ψmidday, which had increased less
than 1 MPa seven days after rewatering. Neither Ψmidday nor
Ψpredawn of previously water stressed N. menziesii seedlings
reached control values by 21 days after rewatering.
The responses of A and gsw to rewatering were slower than
that of Ψxylem. In N. solandri, there was a rapid recovery in A,
whereas in N. menziesii, the recovery was slow. In both species, the recovery in gsw was slow, although it was slightly
faster in N. solandri than in N. menziesii. Throughout the
21-day period after rewatering, gsw of previously water stressed
seedlings of N. solandri did not reach control values.
Response to waterlogging
Figure 2. Changes in xylem water potential (Ψxylem) in response to
decreasing soil matric potential (Ψsoil ).
to decreasing Ψpredawn but approached zero when Ψpredawn
reached about − 4 MPa, which coincided with leaf desiccation.
Regression analyses showed that, in N. solandri and
N. menziesii, A was highly correlated with both the stomatal
component, gsc, and the nonstomatal component, grc, of leaf
conductance (Figure 4). However, the two species differed
significantly in the response of A to gsc (P ≤ 0.05) and of A to
grc (P ≤ 0.01). The slope b of the response curves for N. solandri was 20% greater than for N. menziesii (Table 5). In both
One day after waterlogging treatments were imposed, there
were slight declines in A, gsw and grc in seedlings of both
species (Figure 6), but the effects were not significant until
after 8 days of severe waterlogging. Both A and grc were about
60% less in severely waterlogged seedlings than in control
seedlings of N. solandri (average decreases in A and grc were
from 10.4 to 4.1 µmol m −2 s −1, and from 44.8 to 18.4 mmol
m −2 s −1, respectively). The difference was more than 65% for
N. menziesii (7.4 to 2.3 µmol m −2 s −1 for A and 29.3 to 9.7
mmol m −2 s −1 for grc). In response to waterlogging, gsw declined by 70% in N. solandri (from 216 to 66 mmol m −2 s −1)
and by almost 80% in N. menziesii (from 215 to 50 mmol m −2
s −1).
After 22 days of severe waterlogging, A, gsw and grc fell to
nearly zero in N. menziesii, but severely waterlogged N. solandri seedlings retained some photosynthetic activity for 40
days. In N. menziesii, partial waterlogging resulted in significant (P ≤ 0.05) reductions in gsw until Day 15 and in A on Day
15, but both gsw and A started to recover thereafter. Partial
waterlogging did not alter the water relations of N. solandri.
Under severely waterlogged conditions, N. menziesii seedlings suffered lethal leaf desiccation after 22 days, whereas leaf
desiccation did not occur in N. solandri even after 40 days. No
adventitious roots were observed in seedlings of either species
in the severe waterlogging treatment.
Discussion
Although plants clearly differ in their response to and tolerance
of water stress, it is difficult to define water stress tolerance
quantitatively. The permanent wilting point used in conven-
WATER STRESS AND WATERLOGGING IN NOTHOFAGUS
633
Figure 3. Responses of net photosynthesis
(A), stomatal conductance to diffusion of
water vapor (gsw), and residual conductance (grc) to decreasing predawn xylem
water potential (Ψpredawn).
tional studies of water relations is difficult to define precisely,
and it may not be applicable to many woody plants, such as
N. solandri and N. menziesii, that do not show visible wilting
even under severe water stress because of rigid leaf structures.
We have quantified tolerance of both soil and plant water stress
in N. solandri and N. menziesii from estimates of Ψcri and Ψlethal
, and using these estimates, we have been able to differentiate
between the two species in the water stress tolerance of seedlings grown in confined soil volumes. Although extension of
our results to field-grown trees is not straightforward, differences in water stress tolerance between N. solandri and
N. menziesii determined using Ψcri and Ψlethal appear to be
consistent with ecological observations (Wardle 1970, 1984).
We observed large differences in water stress tolerance
between N. solandri and N. menziesii. Nothofagus solandri
had Ψcri and Ψlethal values of −1.2 and − 7 MPa, respectively,
whereas N. menziesii had values of only −0.7 and −4 MPa,
respectively. Nothofagus solandri has abundant sclerenchyma
associated with highly developed vascular bundles and lignified epidermal cells (Sun 1993), which may explain why it is
more water stress tolerant than N. menziesii, which has structures typical of mesophytic species.
Values of A declined with decreasing Ψxylem in both N. solandri and N. menziesii, but the slope of the relationship
differed significantly. Nothofagus solandri showed a rapid
decrease in A in response to an initial decrease in Ψpredawn, but
subsequently maintained a low but positive value of A as
Ψpredawn fell to −7 MPa. The responses of A and gsw in N. menziesii were initially less steep than in N. solandri, but as Ψpredawn
fell below −4 MPa, the values for both A and gsw approached
zero.
The reduction in A in response to water stress was correlated
634
SUN, SWEET, WHITEHEAD AND BUCHAN
Table 4. Parameters obtained from regression analyses using Equation 1 for net photosynthesis (A), stomatal conductance to diffusion of
water vapor (gsw), and residual conductance to diffusion of CO2 (grc)
in response to predawn xylem water potential (ln|Ψpredawn|). The
values shown are means ± standard errors.
Variable
Parameter
N. solandri
N. menziesii
ln(A)
a
b
r2
a
b
r2
a
b
r2
1.46 ± 0.07
1.23 ± 0.08*1
0.62
4.02 ± 0.05
0.96 ± 0.06
0.68
2.93 ± 0.08
1.26 ± 0.10*
0.55
1.19 ± 0.05
0.78 ± 0.05
0.62
3.54 ± 0.04
0.83 ± 0.05
0.68
2.81 ± 0.06
0.69 ± 0.07
0.40
ln(gsw)
ln(grc)
1
An asterisk indicates that the difference between species is significant at P ≤ 0.001.
with reductions in both stomatal and residual leaf conductance
in both N. solandri and N. menziesii. The impact of stomatal
and nonstomatal inhibition on A depends on species, the severity of the water stress, and the rate of water stress development.
Although stomatal closure may account for the reduction in A
under moderate water stress (Chaves 1991), the significance of
nonstomatal inhibition in water-stressed plants has been
shown in many studies (Mooney et al. 1977, Comstock and
Ehleringer 1984, Ni and Pallardy 1991). Farquhar and Sharkey
(1982) suggested that, although stomatal closure reduces water
loss, it reduces CO2 fixation only marginally and indirectly,
because the latter is limited by the same factors that cause
stomatal closure. Teskey et al. (1986) concluded that internal
limitations are the major cause of reduction of CO2 assimilation in Pinus taeda L., although stomatal response of the
species to several environmental variables is closely correlated
with changes in photosynthetic rate. In water-stressed N. solandri and N. menziesii seedlings, we found that A was more
closely correlated with grc than with gsc.
The rapid response of A to water stress in N. solandri may
have resulted from an adaptation to drought. As Ψpredawn fell
from −0.4 to about −0.8 MPa, there was a sharp decline in gsw
and A in N. solandri seedlings. Such a response may minimize
water loss and possibly increase water stress tolerance. The
high water stress tolerance of N. solandri seedlings was demonstrated by their sustained photosynthetic activity at values of
Ψpredawn as low as −7 MPa, a level that has been observed to
kill many woody plants native to New Zealand (Innes and
Kelly 1992). Photosynthesis and gsw in N. menziesii were less
responsive to water stress than in N. solandri at Ψpredawn values
between −0.2 and −1.0 MPa. Subsequently, however, A declined linearly with decreasing Ψpredawn in both species. Sensi-
Figure 4. Response of net photosynthesis
(A) to stomatal conductance (gsc) and residual conductance (grc) of water-stressed
seedlings.
WATER STRESS AND WATERLOGGING IN NOTHOFAGUS
635
Table 5. Linear regressions and correlations between net photosynthesis (A, µmol m − 2 s − 1) and stomatal conductance to diffusion of CO2 (gsc,
mmol m − 2 s − 1) and residual conductance to diffusion of CO2 (grc, mmol m −2 s −1). The values shown are means ± standard errors.
Dependent variable
Independent variable
Parameter
N. solandri
N. menziesii
A
gsc
a
b
r2
0.94 ± 0.36
0.12 ± 0.006*1
0.76
1.66 ± 0.24
0.10 ± 0.005
0.74
A
grc
a
b
r2
0.73 ± 0.25
0.20 ± 0.006**
0.88
1.17 ± 0.33
0.16 ± 0.01
0.64
1
An asterisk indicates that the difference between species is significant at P ≤ 0.05, and two asterisks indicate that the difference is significant at
P ≤ 0.01.
Figure 5. Recovery in xylem water potential (Ψxylem), net photosynthesis (A), and
stomatal conductance (gsw) following rehydration. Vertical bars indicate standard errors of means of five measurements.
tivity of the response of gas exchange characteristics to water
stress may thus reflect adaptation to drought.
Rewatering severely water-stressed seedlings led to rapid
recovery of both A and Ψxylem in N. solandri, whereas recovery
occurred more slowly in N. menziesii. In both species, the
recovery in A and gsw lagged behind the recovery in Ψxylem,
636
SUN, SWEET, WHITEHEAD AND BUCHAN
Figure 6. Changes in net photosynthesis
(A), stomatal conductance (gsw), and residual conductance (grc) in response to waterlogging. Vertical bars indicate standard
errors of means of five measurements.
indicating that some injury might have occurred to the photosynthetic apparatus during the water stress treatments, possibly increasing susceptibility to photoinhibition (Björkman et
al. 1981, Araus and Hogan 1994). Lack of complete recovery
in Ψmidday was probably due to physical changes such as dying
of root tips, cavitation and embolism of xylem vessels that
adversely affect the movement of water from soil to leaves
(Schulze and Hall 1982, Zimmermann and Milburn 1982,
Tyree and Sperry 1989). Nothofagus menziesii was damaged
more than N. solandri under conditions of severe water stress,
as indicated by the poor recovery of Ψpredawn, despite the
increased soil water availability. The response of Ψxylem to
decreasing Ψsoil in N. solandri and its tolerance to water stress
suggests that the species may be classified as dehydration
tolerant, whereas N. menziesii exhibited dehydration postponement; that is, this species is capable of maintaining relatively high water potential under conditions of moderate water
stress.
Tolerance to waterlogging is correlated with changes in A,
gsw and transpiration in many woody plants (Regehr et al.
1975, Pereira and Kozlowski 1977, Zaerr 1983, van der
Moezel et al. 1989). Stomatal closure usually occurs within a
few days after the soil is waterlogged, and in many species
sensitive to waterlogging, the stomata remain closed for a long
time (Kozlowski 1984). In some species tolerant to waterlogging, stomata may reopen after a critical period. In Fraxinus
pennsylvanica Marsh. seedlings, stomata close rapidly following waterlogging but reopen after about 2 weeks (Sena Gomes
and Kozlowski 1980). Stomata of Eucalyptus camaldulensis
Dehnh. reopen about 5 weeks after waterlogging (van der
Moezel et al. 1989). Tang and Kozlowski (1982) reported that
stomata of Quercus macrocarpa Michx. seedlings reopen
about 30 days after flooding. Reopening of stomata did not
occur in severely waterlogged N. solandri or N. menziesii
seedlings during the experiment.
Nothofagus solandri and N. menziesii differed in sensitivity
WATER STRESS AND WATERLOGGING IN NOTHOFAGUS
to waterlogging. Moderate waterlogging resulted in a reduction of gas exchange properties in N. menziesii, whereas N. solandri seedlings were not affected. However, neither species
was well adapted to severe or prolonged waterlogging. Severe
waterlogging reduced A, gsw and grc in both N. solandri and
N. menziesii, but the patterns of reduction differed between the
two species. Regehr et al. (1975) observed a 50% reduction in
A in waterlogging-tolerant Populus deltoides Bartr. ex Marsh.
during complete inundation of the root system for 28 days. In
our study, A was reduced by approximately 50% in N. solandri
and by more than 65% in N. menziesii after only 8 days of
waterlogging. In moderately waterlogging-tolerant Liquidambar styraciflua L. seedlings, A was reduced by 70% over a
9-day flooding period (Pezeshki and Chambers 1985). Zaerr
(1983) showed that following flooding, A decreased to about
half the control rate within 4 days for Pseudotsga menziesii
(Mirb.) Franco and Picea abies (L.) Karst., but did not change
in Pinus sylvestris L. In N. solandri and N. menziesii, the
reduction in A of seedlings under waterlogged conditions was
associated with decreased gsw and with a marked reduction in
grc, indicating that waterlogging resulted in severe damage to
the photosynthetic apparatus of the plants.
In New Zealand, N. solandri is adapted to a wider range of
soil water conditions than N. menziesii. Nothofagus solandri is
usually either the sole or the dominant species in forests of dry
habitats or on poorly drained soils (Wardle 1970, 1984), and it
codominates with N. menziesii at the wetter end of its ecological range. In areas where the climate is dry and where both
species are present, N. solandri is most likely to occupy the
exposed sites, whereas N. menziesii is more or less restricted
to sheltered sites (Wardle 1984). Our results are consistent with
these observations.
Acknowledgments
The authors thank Karl Schasching for his help in seed collection and
for organizing research facilities. We also thank B. Bullsmith, D.
Clark, D. Conder and P. Fuller for technical assistance, and T. Pearson
for help with preparing graphs. O.J. Sun is grateful for the financial
support provided by T.W. Adams Scholarship, BNZ Scholarship for
Forestry Research, Robert C. Bruce Trust and Young Scientists’ Fund
of the Royal Society of New Zealand, and support of NZ Forest
Research Institute in preparing the manuscript. Comments from K.P.
Hogan on the manuscript are greatly appreciated.
References
Araus, J.L. and K.P. Hogan. 1994. Leaf structure and patterns of
photoinhibition in two neotropical palms in clearings and forest
understory during the dry season. Am. J. Bot. 81:726--738.
Björkman, O., S.B. Powles, D.C. Fork and G. Öquist. 1981. Interaction between high irradiance and water stress on photosynthetic
reactions. Year Book, Carnegie Inst., Washington, D.C. 80:57--59.
Buchan, G.D. and K.S. Grewal. 1990. The power-function models for
the soil moisture characteristics. J. Soil Sci. 41:111--117.
Chaves, M.M. 1991. Effects of water deficits on carbon assimilation.
J. Exp. Bot. 42:1--16.
Comstock, J. and J. Ehleringer. 1984. Photosynthetic response to
slowly decreasing leaf water potential in Encelia frutescens. Oecologia 61:241--248.
637
Farquhar, G.D. and T.D. Sharkey. 1982. Stomatal conductance and
photosynthesis. Annu. Rev. Plant Physiol. 33:317--345.
Hanks, S.L. and D.E. Fairbrothers. 1976. Palynotaxonomic investigation of Fagus L. and Nothofagus BL. Light microscopy, scanning
electron microscopy, and computer analysis. In Botanical Systematics 1. Ed. V.H. Heywood. Academic Press, London, U.K.,
pp 11--141.
Hill, R.S. and J. Read. 1991. A revised infrageneric classification of
Nothofagus (Fagaceae). Bot. J. Linn. Soc. 105:37--72.
Ingestad, T. 1971. A definition of optimum nutrient requirements in
birch seedlings. II. Physiol. Plant. 24:118--125.
Innes, K.P. and D. Kelly. 1992. Water potentials in native woody
vegetation during and after a drought in Canterbury. N.Z. J. Bot.
30:81--94.
Jane, G.T. and T.G.A. Green. 1985. Patterns of stomatal conductance
in six evergreen tree species from a New Zealand cloud forest. Bot.
Gaz. 146:413--420.
Jane, G.T. and T.G.A. Green. 1986. Etiology of forest dieback areas
within the Kaimai Range, North Island, New Zealand. N.Z. J. Bot.
24:513--527.
Janiesch, P. 1991. Ecophysiological adaptations of higher plants in
natural communities to waterlogging. In Ecological Responses to
Environmental Stresses. Eds. J. Rozema and J.A.C. Verkleij. Kluwer Academic Publishers, The Netherlands, pp 50--60.
Kozlowski, T.T. 1984. Plant responses to flooding of soil. BioScience
34:162--167.
McGlone, M.S. 1985. Plant biogeography and the late Cenozoic history of New Zealand. N.Z. J. Bot. 23:723--749.
Mildenhall, D.C. 1980. New Zealand late Cretaceous and Cenozoic
plant biogeography: a contribution. Palaeogeogr. Palaeoclimatol.
Palaeoecol. 31:197--233.
Mooney, H.A., O. Björkman and G.J. Collatz. 1977. Photosynthetic
acclimation to temperature and drought in the desert shrub Larrea
divarcata. Year Book, Carnegie Inst., Washington, D.C. 76:328--335.
Myers, B.J. and J.J. Landsberg. 1989. Drought and seedling growth of
two eucalypt species from contrasting habitats. Tree Physiol.
5:207--218.
Ni, B.-R. and S.G. Pallardy. 1991. Response of gas exchange to water
stress in seedlings of woody angiosperms. Tree Physiol. 8:1--9.
Passioura, J.B. 1982. Water in the soil--plant--atmosphere continuum.
In Physiological Plant Ecology II. Water Relations and Carbon
Assimilation. Encyclopedia of Plant Physiology, New Series, Vol.
12B. Eds. O.L. Lange, P.S. Nobel, C.B. Osmond and H. Ziegler.
Springer-Verlag, Berlin, Heidelberg, pp 5--33.
Pereira, J.S. and T.T. Kozlowski. 1977. Variations among woody
angiosperms in relation to flooding. Physiol. Plant. 41:184--192.
Pezeshki, S.R. and J.L. Chambers. 1985. Stomatal and photosynthesis
response of sweet gum Liquidambar styraciflua to flooding. Can. J.
For. Res. 15:371--375.
Philipson, W.R. and M.N. Philipson. 1988. A classification of the
genus Nothofagus (Fagaceae). Bot. J. Linn. Soc. 98:27--36.
Ranney, T.G., T.H. Whitlow and N.L. Bassuk. 1990. Response of five
temperate deciduous tree species to water stress. Tree Physiol.
6:439--448.
Regehr, D.L., F.A. Bazzaz and W.R. Boggess. 1975. Photosynthesis,
transpiration, and leaf conductance of Populus deltoides in relation
to flooding and drought. Photosynthetica 9:52--61.
Schulze, E.-D. and A.E. Hall. 1982. Stomatal responses, water loss
and CO2 assimilation rates of plants in contrasting environments. In
Physiological Plant Ecology II. Water Relations and Carbon Assimilation. Encyclopedia of Plant Physiology, New Series, Vol.
12B. Eds. O.L. Lange, P.S. Nobel, C.B. Osmond and H. Ziegler.
Springer-Verlag, Berlin, Heidelberg, pp 181--230.
638
SUN, SWEET, WHITEHEAD AND BUCHAN
Seiler, J.R. and B.H. Cazell. 1990. Influence of water stress on the
physiology and growth of red spruce seedlings. Tree Physiol. 6:69-77.
Sena Gomes, A.R. and T.T. Kozlowski. 1980. Growth response and
adaptation of Fraxinus pennsylvania seedlings to flooding. Plant
Physiol. 66:267--271.
Sun, O.J. 1993. Genetics and physiology of Nothofagus solandri var.
cliffortioides and N. menziesii----a comparative study. Ph.D. Thesis.
University of Canterbury, Christchurch, New Zealand, 279 p.
Tang, Z.C. and T.T. Kozlowski. 1982. Some physiological and morphological responses of Quercus macrocarpa seedlings to flooding.
Can. J. For. Res. 12:196--202.
Teskey, R.O., J.A. Fites, L.J. Samuelson and B.C. Bongarten. 1986.
Stomatal and nonstomatal limitations to net photosynthesis in Pinus
taeda L. under different environmental conditions. Tree Physiol.
2:131--142.
Tyree, M.T., J. Alexander and J.-L. Machado. 1992. Loss of hydraulic
conductivity due to drought in intact juveniles of Quercus rubra and
Populus deltoides. Tree Physiol. 10:411--415.
Tyree, M.T. and J.S. Sperry. 1989. Vulnerability of xylem to cavitation
and embolism. Annu. Rev. Plant Physiol. Mol. Biol. 40:19--38.
van der Moezel, P.G., L.E. Watson and D.T. Bell. 1989. Gas exchange
responses of two Eucalyptus species to salinity and waterlogging.
Tree Physiol. 5:251--257.
Wardle, J. 1970. The ecology of Nothofagus solandri. 1. The distribution and relationship with other major forest and shrub species. N.Z.
J. Bot. 8:494--531.
Wardle, J. 1984. The New Zealand beeches. Ecology, utilisation and
management. New Zealand Forest Service, Christchurch, New Zealand, 447 p.
Wardle, P. 1967. Biological flora of New Zealand. 2. Nothofagus
menziesii (Hook. f.) Oerst. (Fagaceae). N.Z. J. Bot. 5:276--302.
Zaerr, J.B. 1983. Short-term flooding and net photosynthesis in seedlings of three conifers. For. Sci. 29:71--78.
Zimmermann, U. and J.A. Milburn. 1982. Transport and storage of
water. In Physiological Plant Ecology II. Water Relations and Carbon Assimilation. Encyclopedia of Plant Physiology, New Series Vol.
12B. Eds. O.L. Lange, P.S. Nobel, C.B. Osmond and H. Ziegler.
Springer-Verlag, Berlin, Heidelberg, pp 135--151.
© 1995 Heron Publishing----Victoria, Canada
Physiological responses to water stress and waterlogging in Nothofagus
species
OSBERT J. SUN,1,2 GEOFFREY B. SWEET,1 DAVID WHITEHEAD3 and GRAEME D.
BUCHAN4
1
2
3
4
School of Forestry, University of Canterbury, Private Bag 4800, Christchurch, New Zealand
Present address: New Zealand Forest Research Institute, Private Bag 4800, Christchurch, New Zealand
Manaaki Whenua--Landcare Research, P.O. Box 69, Lincoln, Canterbury, New Zealand
Soil Science Department, Lincoln University, Canterbury, New Zealand
Received December 13, 1994
Summary Gas exchange and water relations were investigated in Nothofagus solandri var. cliffortioides (Hook. f.) Poole
(mountain beech) and Nothofagus menziesii (Hook. f.) Oerst
(silver beech) seedlings in response to water stress and waterlogging. At soil matric potentials (Ψsoil ) above − 0.005 MPa,
N. solandri had significantly higher photosynthetic rates (A),
and stomatal and residual conductances (gsw and grc), and lower
predawn xylem water potentials (Ψpredawn) than N. menziesii.
The relative tolerance of plants to water stress was defined in
terms of critical soil matric potential (Ψcri ) and lethal xylem
water potential (Ψlethal). The estimated values of Ψcri and Ψlethal
were −1.2 and − 7 MPa, respectively, for N. solandri, and −0.7
and −4 MPa, respectively, for N. menziesii. Photosynthesis was
sustained to a xylem water potential (Ψxylem) of − 7 MPa in
N. solandri compared with − 4 MPa in N. menziesii.
Following rewatering, both A and Ψxylem recovered quickly
in N. solandri, whereas the two variables recovered more
slowly in N. menziesii. During the development of water stress,
nonstomatal inhibition significantly affected A in both N. solandri and N. menziesii. Nothofagus menziesii was more susceptible to inhibition of A by waterlogging than N. solandri.
However, the tolerance of N. solandri to severe waterlogging
was also limited as a result of a failure to form adventitious
roots, suggesting a lack of adaptation to these conditions. The
differences in tolerance to water stress and waterlogging between the two species are consistent with the distribution
patterns of N. solandri and N. menziesii in New Zealand.
Keywords: gas exchange, geographical distribution, nonstomatal inhibition, Nothofagus menziesii, Nothofagus solandri
var. cliffortioides, water relations, water stress.
Introduction
Woody plants vary markedly in their response to and tolerance
of water stress (e.g., Myers and Landsberg 1989, Ranney et al.
1990, Seiler and Cazell 1990, Ni and Pallardy 1991) and
waterlogging (e.g., Pereira and Kozlowski 1977, Sena Gomes
and Kozlowski 1980, Tang and Kozlowski 1982, Pezeshki and
Chambers 1985). Water stress influences metabolism, physiology and morphology in plants (Passioura 1982). Waterlogging
causes inadequate aeration in soil, leading to rapid depletion
of oxygen which induces many physiological and morphological changes (Kozlowski 1984), and affects mineralization and
solubility of mineral substances, and leads to the formation of
phytotoxic compounds (Janiesch 1991).
Nothofagus solandri var. cliffortioides (Hook. f.) Poole and
Nothofagus menziesii (Hook. f.) Oerst are two of the five taxa
in the genus Nothofagus (Southern beech) that are native to
New Zealand. They both have a wide ecological and geographical distribution, occurring from sea level to the upper
timberlines at comparable latitudes. Taxonomically, however,
the two species belong to two subgroups with distinct pollen
types (Hanks and Fairbrothers 1976, Philipson and Philipson
1988, Hill and Read 1991). The genetic divergence of the two
species may be dated back to the upper Cretaceous (Wardle
1984), approximately 70 million years ago, before New Zealand was separated from Australia and Antarctica (Mildenhall
1980, McGlone 1985). From the present distribution, it appears that soil water availability and drainage may play a
prominent role in differentiating the geographical occurrence
of N. solandri and N. menziesii (Wardle 1984). Drought has
been suggested as an important factor in forest mortality that
has led to the present vegetation pattern in New Zealand (Jane
and Green 1986, Innes and Kelly 1992). Susceptibility of trees
to waterlogging followed by drought has also been linked to
forest decline (Jane and Green 1986). Plants that have been
waterlogged may be more sensitive to drought than plants
growing in aerated soils as a consequence of reductions in size
of the root systems in response to inundated soil conditions
(Jane and Green 1985, 1986). Nothofagus solandri is usually
more dominant than N. menziesii in habitats with a dry climate
(Wardle 1984), whereas on waterlogged soils, N. menziesii is
stunted and often replaced by species such as N. solandri and
Podocarpus dacrydioides A. Rich. (Wardle 1967). Despite
many studies of the ecology and ecosystem dynamics of Nothofagus-dominated forests in New Zealand, a detailed knowledge of water relations of Nothofagus under conditions of
630
SUN, SWEET, WHITEHEAD AND BUCHAN
water stress and waterlogging is lacking. To investigate the role
of water relations in determining the geographical distribution
of N. solandri and N. menziesii in New Zealand, we tested (1)
the responses of photosynthesis and xylem water potential to
soil water availability, (2) tolerance of seedlings to progressively increased soil drying, and (3) tolerance to waterlogging
in the two species.
Materials and methods
Response to water stress
Plant material, experimental design and treatment In the
spring of the first growing season, seedlings were raised from
seed and grown in pots containing a mix of peat (60%), soils
collected from a Nothofagus-dominated forest (20%), ground
pine bark (10%), coarse sand (10%) and slow-release fertilizer
(2 kg m −3, Osmocote® Plus, Grace-Sierra, Heerlen, The Netherlands). On March 20 (autumn) of the second growing season,
15 seedlings of each species were each transplanted to a 4-liter
pot containing nursery top soil (loam-textured) packed at a bulk
density (ρb) of approximately 1100 kg m −3. Leaf area (onesided) was estimated as 0.06 m2 for N. solandri and 0.05 m2 for
N. menziesii seedlings. The seedlings were arranged on a bench
in a greenhouse, and a dilute complete nutrient solution
(Ingestad 1971) containing 10 ppm nitrogen was applied regularly until treatments began. The photoperiod in the greenhouse
was extended to 16 h with 400 W sodium vapor lamps. Maximum photosynthetic photon flux density (Q) varied between
1000 µmol m −2 s −1 in midsummer and 600 µmol m −2 s −1 in
winter. Day/night temperatures (25/15 °C) were controlled
thermostatically.
Treatments began 4 weeks after transplanting. The experiment comprised three water stress treatments in a completely
randomized single-tree plot design, and each treatment contained five seedlings of each species. Seedlings in Treatment 1
were well watered throughout the experiment (control), to
maintain the soil matric potential (Ψsoil ) above −0.005 MPa.
Seedlings in Treatment 2 were rewatered on Day 30 when Ψsoil
reached a critical level (see below). Seedlings in Treatment 3
Figure 1. Relationship between matric potential (Ψsoil ) and volumetric
water content (θv) of the soil.
were left unwatered for 52 days until the end of the experiment.
All pots were placed randomly on the bench and rearranged
every second day.
Determination of soil matric potential Soil matric potential
(Ψsoil ) was estimated from volumetric soil water content (θv)
based on an equation describing the soil water characteristic
(Figure 1) determined using a tension table and pressure plate
apparatus (Buchan and Grewal 1990). Measurements of θv
were made with a time domain reflectometry (TDR) soil moisture meter (SoilMoisture Equipment Corp., Santa Barbara, CA,
USA).
Assessment of critical soil matric and plant water potentials
Tolerance to water stress was defined in terms of critical soil
matric potential (Ψcri ) and lethal xylem water potential (Ψlethal).
The value of Ψsoil at which Ψpredawn equals Ψmidday (of the
preceding day) is defined as Ψcri , i.e., the minimum Ψsoil for
adequate water uptake. When a plant reaches Ψcri , the hydraulic conductivity from the soil to leaves decreases abruptly
(Tyree et al. 1992), so that continuing dehydration causes Ψxylem
to decrease until leaf desiccation occurs. Survival and recovery
of the plant following rewatering may then depend on the
plant’s ability to tolerate Ψxylem that induces leaf desiccation.
The values of Ψxylem below which leaf desiccation occurred
were defined as Ψlethal.
Values of Ψcri for the two species were determined graphically from measurements of Ψpredawn and Ψmidday, and values of
Ψlethal were taken as the minimum Ψxylem before leaf desiccation became visible.
Measurements of photosynthesis and leaf conductance
Stomatal conductance to diffusion of water vapor (gsw), rates
of net photosynthesis (A) and intercellular CO2 concentration
(Ci) were measured at ambient CO2 concentration (about 380
µmol mol −1) in a controlled environment cabinet (Temperzone,
Auckland, New Zealand) between 1100 and 1400 h (NZ ST)
with a portable photosynthesis system (LI-6200, Li-Cor Inc.,
Lincoln, NE, USA). Measurements were made on twigs with
5--10 fully expanded leaves.
Seedlings were transferred from the greenhouse to the controlled environment cabinet at least 2 h before measurements
were made. Photosynthetic photon flux density (Q) in the
cabinet was 650 µmol m −2 s −1, and temperature and relative
humidity were maintained at 20 °C and 55%, respectively.
After each measurement, twigs were detached and immediately used for Ψmidday measurements. Leaf areas (one-sided)
were measured with an image area meter (Delta-T Devices,
Burwell, Cambridge, U.K.).
Leaf conductance was partitioned into stomatal (gsc) and
nonstomatal or residual conductance to diffusion of CO2 (grc)
according to Farquhar and Sharkey (1982), where the term grc
includes both mesophyll conductance (gmc) and carboxylation
efficiency (k).
Measurements of xylem water potential Measurements of
Ψxylem were made with a pressure chamber (PMS Instrument
Co., Corvallis, OR, USA). Measurements of Ψmidday were made
WATER STRESS AND WATERLOGGING IN NOTHOFAGUS
immediately after gas exchange measurements in the controlled environment cabinet between 1100 and 1400 h, and Ψpredawn
was measured in the greenhouse on the following day, about
14 h later. Twigs for measuring Ψxylem were obtained from a
second- or third-order branch in the middle of the main stem.
Data analysis Estimates of A, gsw, grc, Ψmidday and Ψpredawn in
Treatment 1 were made over a 52-day period. Results were
pooled for each species and compared by the Student’s t-test.
Values of Ψcri and Ψlethal were well separated between species
and were not compared statistically. Data for Treatments 1 and
3, and data from the early part of Treatment 2 (before soils were
rewatered) were used to determine the relationship between
Ψpredawn and Ψsoil , responses of A, gsw and grc to Ψpredawn, and
stomatal and nonstomatal (residual) components of photosynthesis. The data were evaluated by regression analysis and
analysis of variance.
631
test with a confidence level of P ≤ 0.05. Conditions in the
greenhouse on the measurement days are given in Table 2.
Results
Response to water stress
When Ψsoil was maintained above −0.005 MPa, A, gsw and grc
were significantly (P ≤ 0.001) greater by 30, 25 and 30%,
respectively, and Ψpredawn significantly lower by 30% for N. solandri than for N. menziesii (Table 3). However, the values for
Ψmidday did not differ between the species.
The effect of decreasing Ψsoil on Ψxylem differed according
to species (Figure 2). In N. solandri, decreases in both Ψmidday
and Ψpredawn occurred gradually in response to declining Ψsoil ,
whereas in N. menziesii, Ψmidday and Ψpredawn initially decreased slowly with decreasing Ψsoil but declined abruptly
when Ψsoil fell below −0.5 MPa. Values of Ψcri were estimated
as −1.2 and −0.7 MPa, and values of Ψlethal were estimated as
− 7 and − 4 MPa for N. solandri and N. menziesii, respectively.
In both species, values of A, gsw and grc declined with
decreasing Ψpredawn (Figure 3) following a relationship of the
form:
Response to waterlogging
Fifteen seedlings of each species were grown in pots containing potting mix as described previously. Soil water content of
the control group was maintained near field capacity throughout the experiment. The water table in the second group of
seedlings was raised 50 to 60 mm by placing pots in saucers
containing tap water (partial waterlogging). Seedlings in the
third group were totally flooded by submerging the pots individually in water containers (severe waterlogging). The water
in the containers was replaced daily with fresh tap water. All
pots were placed randomly on a bench in the greenhouse and
rearranged weekly. The air-filled porosity (εa) for the partial
waterlogging treatment was estimated to be only half of that of
the control (Table 1).
Sequential measurements of A and gsw were conducted on
twigs at ambient CO2 concentration between 1200 and 1400 h
in the greenhouse, and effects were evaluated by analysis of
variance. Means were compared by Duncan’s multiple-range
ln (y) = a − b ln|Ψ predawn |,
(1)
where b defines the sensitivity of the response. Values for
parameters a and b are summarized in Table 4. Both A and grc
were more sensitive to decreasing Ψpredawn (P ≤ 0.001) in
N. solandri than in N. menziesii, but there was no significant
difference for gsw.
Rapid reductions in A, gsw and grc occurred as Ψpredawn fell
from −0.4 to −0.8 MPa in N. solandri, and from −0.2 to −1.0
MPa in N. menziesii. In N. solandri, the reductions in A, gsw
and grc continued until Ψpredawn fell to − 7 MPa, whereas in
N. menziesii, A, gsw and grc initially declined slowly in response
Table 1. Soil bulk density (ρb), total porosity (ε), volumetric soil water content (θv), and air-filled porosity (εa) for three waterlogging treatments.
The values shown are means ± standard errors, n = 6.
Treatment
ρb (kg m − 3)
ε (%)
θv (%)
εa (%)
Control
Partial waterlogging
Severe waterlogging
776 ± 11
70.7 ± 0.04
50.2 ± 1.9
59.0 ± 0.3
70.7
20.5 ± 0.5
11.7 ± 0.7
0
Table 2. Mean photosynthetic photon flux density (Q), mean air temperature (Tair ), mean ambient CO2 concentration (Ca), and mean vapor pressure
deficit (D) in the greenhouse during the experimental period.
Days since start
of treatment
Q
(µmol m −2 s −1)
Tair
(°C)
Ca
(µmol mol − 1)
D
(kPa)
1
8
15
22
40
352
840
930
840
914
24.0
26.8
25.8
26.1
26.7
341
340
344
335
348
1.66
1.98
2.38
2.05
1.84
632
SUN, SWEET, WHITEHEAD AND BUCHAN
Table 3. Net photosynthetic rates (A), stomatal conductance to diffusion of water vapor (gsw), residual conductance to diffusion of CO2
(grc), and midday and predawn xylem water potentials (Ψmidday and
Ψpredawn) of seedlings at Ψsoil ≥ −0.005 MPa. The values shown are
means ± standard errors.
Variable
N. solandri
N. menziesii
A (µmol m −2 s −1)
gsw (mmol m −2 s −1)
grc (mmol m − 2 s − 1)
Ψmidday (MPa)
Ψpredawn(MPa)
10.61 ± 0.34*1
123.8 ± 4.0*
46.5 ± 2.0*
−1.15 ± 0.02
−0.52 ± 0.01*
8.17 ± 0.36
99.1 ± 5.2
34.5 ± 1.7
−1.14 ± 0.03
−0.35 ± 0.02
1
An asterisk indicates that the difference between species is significant at P ≤ 0.001.
species, b for the response of A to grc was more than 60%
greater than for the response of A to gsc.
When seedlings were rewatered after being subjected to
severe water stress (Ψcri ≤ −1.2 and −0.7 MPa for N. solandri
and N. menziesii, respectively), Ψxylem, A and gsw increased
more rapidly for N. solandri seedlings than for N. menziesii
seedlings (Figure 5). Three days after rewatering, both Ψmidday
and Ψpredawn had recovered by more than 2 MPa in N. solandri,
and Ψpredawn recovered to the control values after 14 days.
However, in rewatered seedlings that had been water stressed
previously, Ψmidday remained consistently lower than in control
seedlings throughout the experiment. In N. menziesii, the recovery of Ψxylem following rewatering was slower than in
N. solandri, especially for Ψmidday, which had increased less
than 1 MPa seven days after rewatering. Neither Ψmidday nor
Ψpredawn of previously water stressed N. menziesii seedlings
reached control values by 21 days after rewatering.
The responses of A and gsw to rewatering were slower than
that of Ψxylem. In N. solandri, there was a rapid recovery in A,
whereas in N. menziesii, the recovery was slow. In both species, the recovery in gsw was slow, although it was slightly
faster in N. solandri than in N. menziesii. Throughout the
21-day period after rewatering, gsw of previously water stressed
seedlings of N. solandri did not reach control values.
Response to waterlogging
Figure 2. Changes in xylem water potential (Ψxylem) in response to
decreasing soil matric potential (Ψsoil ).
to decreasing Ψpredawn but approached zero when Ψpredawn
reached about − 4 MPa, which coincided with leaf desiccation.
Regression analyses showed that, in N. solandri and
N. menziesii, A was highly correlated with both the stomatal
component, gsc, and the nonstomatal component, grc, of leaf
conductance (Figure 4). However, the two species differed
significantly in the response of A to gsc (P ≤ 0.05) and of A to
grc (P ≤ 0.01). The slope b of the response curves for N. solandri was 20% greater than for N. menziesii (Table 5). In both
One day after waterlogging treatments were imposed, there
were slight declines in A, gsw and grc in seedlings of both
species (Figure 6), but the effects were not significant until
after 8 days of severe waterlogging. Both A and grc were about
60% less in severely waterlogged seedlings than in control
seedlings of N. solandri (average decreases in A and grc were
from 10.4 to 4.1 µmol m −2 s −1, and from 44.8 to 18.4 mmol
m −2 s −1, respectively). The difference was more than 65% for
N. menziesii (7.4 to 2.3 µmol m −2 s −1 for A and 29.3 to 9.7
mmol m −2 s −1 for grc). In response to waterlogging, gsw declined by 70% in N. solandri (from 216 to 66 mmol m −2 s −1)
and by almost 80% in N. menziesii (from 215 to 50 mmol m −2
s −1).
After 22 days of severe waterlogging, A, gsw and grc fell to
nearly zero in N. menziesii, but severely waterlogged N. solandri seedlings retained some photosynthetic activity for 40
days. In N. menziesii, partial waterlogging resulted in significant (P ≤ 0.05) reductions in gsw until Day 15 and in A on Day
15, but both gsw and A started to recover thereafter. Partial
waterlogging did not alter the water relations of N. solandri.
Under severely waterlogged conditions, N. menziesii seedlings suffered lethal leaf desiccation after 22 days, whereas leaf
desiccation did not occur in N. solandri even after 40 days. No
adventitious roots were observed in seedlings of either species
in the severe waterlogging treatment.
Discussion
Although plants clearly differ in their response to and tolerance
of water stress, it is difficult to define water stress tolerance
quantitatively. The permanent wilting point used in conven-
WATER STRESS AND WATERLOGGING IN NOTHOFAGUS
633
Figure 3. Responses of net photosynthesis
(A), stomatal conductance to diffusion of
water vapor (gsw), and residual conductance (grc) to decreasing predawn xylem
water potential (Ψpredawn).
tional studies of water relations is difficult to define precisely,
and it may not be applicable to many woody plants, such as
N. solandri and N. menziesii, that do not show visible wilting
even under severe water stress because of rigid leaf structures.
We have quantified tolerance of both soil and plant water stress
in N. solandri and N. menziesii from estimates of Ψcri and Ψlethal
, and using these estimates, we have been able to differentiate
between the two species in the water stress tolerance of seedlings grown in confined soil volumes. Although extension of
our results to field-grown trees is not straightforward, differences in water stress tolerance between N. solandri and
N. menziesii determined using Ψcri and Ψlethal appear to be
consistent with ecological observations (Wardle 1970, 1984).
We observed large differences in water stress tolerance
between N. solandri and N. menziesii. Nothofagus solandri
had Ψcri and Ψlethal values of −1.2 and − 7 MPa, respectively,
whereas N. menziesii had values of only −0.7 and −4 MPa,
respectively. Nothofagus solandri has abundant sclerenchyma
associated with highly developed vascular bundles and lignified epidermal cells (Sun 1993), which may explain why it is
more water stress tolerant than N. menziesii, which has structures typical of mesophytic species.
Values of A declined with decreasing Ψxylem in both N. solandri and N. menziesii, but the slope of the relationship
differed significantly. Nothofagus solandri showed a rapid
decrease in A in response to an initial decrease in Ψpredawn, but
subsequently maintained a low but positive value of A as
Ψpredawn fell to −7 MPa. The responses of A and gsw in N. menziesii were initially less steep than in N. solandri, but as Ψpredawn
fell below −4 MPa, the values for both A and gsw approached
zero.
The reduction in A in response to water stress was correlated
634
SUN, SWEET, WHITEHEAD AND BUCHAN
Table 4. Parameters obtained from regression analyses using Equation 1 for net photosynthesis (A), stomatal conductance to diffusion of
water vapor (gsw), and residual conductance to diffusion of CO2 (grc)
in response to predawn xylem water potential (ln|Ψpredawn|). The
values shown are means ± standard errors.
Variable
Parameter
N. solandri
N. menziesii
ln(A)
a
b
r2
a
b
r2
a
b
r2
1.46 ± 0.07
1.23 ± 0.08*1
0.62
4.02 ± 0.05
0.96 ± 0.06
0.68
2.93 ± 0.08
1.26 ± 0.10*
0.55
1.19 ± 0.05
0.78 ± 0.05
0.62
3.54 ± 0.04
0.83 ± 0.05
0.68
2.81 ± 0.06
0.69 ± 0.07
0.40
ln(gsw)
ln(grc)
1
An asterisk indicates that the difference between species is significant at P ≤ 0.001.
with reductions in both stomatal and residual leaf conductance
in both N. solandri and N. menziesii. The impact of stomatal
and nonstomatal inhibition on A depends on species, the severity of the water stress, and the rate of water stress development.
Although stomatal closure may account for the reduction in A
under moderate water stress (Chaves 1991), the significance of
nonstomatal inhibition in water-stressed plants has been
shown in many studies (Mooney et al. 1977, Comstock and
Ehleringer 1984, Ni and Pallardy 1991). Farquhar and Sharkey
(1982) suggested that, although stomatal closure reduces water
loss, it reduces CO2 fixation only marginally and indirectly,
because the latter is limited by the same factors that cause
stomatal closure. Teskey et al. (1986) concluded that internal
limitations are the major cause of reduction of CO2 assimilation in Pinus taeda L., although stomatal response of the
species to several environmental variables is closely correlated
with changes in photosynthetic rate. In water-stressed N. solandri and N. menziesii seedlings, we found that A was more
closely correlated with grc than with gsc.
The rapid response of A to water stress in N. solandri may
have resulted from an adaptation to drought. As Ψpredawn fell
from −0.4 to about −0.8 MPa, there was a sharp decline in gsw
and A in N. solandri seedlings. Such a response may minimize
water loss and possibly increase water stress tolerance. The
high water stress tolerance of N. solandri seedlings was demonstrated by their sustained photosynthetic activity at values of
Ψpredawn as low as −7 MPa, a level that has been observed to
kill many woody plants native to New Zealand (Innes and
Kelly 1992). Photosynthesis and gsw in N. menziesii were less
responsive to water stress than in N. solandri at Ψpredawn values
between −0.2 and −1.0 MPa. Subsequently, however, A declined linearly with decreasing Ψpredawn in both species. Sensi-
Figure 4. Response of net photosynthesis
(A) to stomatal conductance (gsc) and residual conductance (grc) of water-stressed
seedlings.
WATER STRESS AND WATERLOGGING IN NOTHOFAGUS
635
Table 5. Linear regressions and correlations between net photosynthesis (A, µmol m − 2 s − 1) and stomatal conductance to diffusion of CO2 (gsc,
mmol m − 2 s − 1) and residual conductance to diffusion of CO2 (grc, mmol m −2 s −1). The values shown are means ± standard errors.
Dependent variable
Independent variable
Parameter
N. solandri
N. menziesii
A
gsc
a
b
r2
0.94 ± 0.36
0.12 ± 0.006*1
0.76
1.66 ± 0.24
0.10 ± 0.005
0.74
A
grc
a
b
r2
0.73 ± 0.25
0.20 ± 0.006**
0.88
1.17 ± 0.33
0.16 ± 0.01
0.64
1
An asterisk indicates that the difference between species is significant at P ≤ 0.05, and two asterisks indicate that the difference is significant at
P ≤ 0.01.
Figure 5. Recovery in xylem water potential (Ψxylem), net photosynthesis (A), and
stomatal conductance (gsw) following rehydration. Vertical bars indicate standard errors of means of five measurements.
tivity of the response of gas exchange characteristics to water
stress may thus reflect adaptation to drought.
Rewatering severely water-stressed seedlings led to rapid
recovery of both A and Ψxylem in N. solandri, whereas recovery
occurred more slowly in N. menziesii. In both species, the
recovery in A and gsw lagged behind the recovery in Ψxylem,
636
SUN, SWEET, WHITEHEAD AND BUCHAN
Figure 6. Changes in net photosynthesis
(A), stomatal conductance (gsw), and residual conductance (grc) in response to waterlogging. Vertical bars indicate standard
errors of means of five measurements.
indicating that some injury might have occurred to the photosynthetic apparatus during the water stress treatments, possibly increasing susceptibility to photoinhibition (Björkman et
al. 1981, Araus and Hogan 1994). Lack of complete recovery
in Ψmidday was probably due to physical changes such as dying
of root tips, cavitation and embolism of xylem vessels that
adversely affect the movement of water from soil to leaves
(Schulze and Hall 1982, Zimmermann and Milburn 1982,
Tyree and Sperry 1989). Nothofagus menziesii was damaged
more than N. solandri under conditions of severe water stress,
as indicated by the poor recovery of Ψpredawn, despite the
increased soil water availability. The response of Ψxylem to
decreasing Ψsoil in N. solandri and its tolerance to water stress
suggests that the species may be classified as dehydration
tolerant, whereas N. menziesii exhibited dehydration postponement; that is, this species is capable of maintaining relatively high water potential under conditions of moderate water
stress.
Tolerance to waterlogging is correlated with changes in A,
gsw and transpiration in many woody plants (Regehr et al.
1975, Pereira and Kozlowski 1977, Zaerr 1983, van der
Moezel et al. 1989). Stomatal closure usually occurs within a
few days after the soil is waterlogged, and in many species
sensitive to waterlogging, the stomata remain closed for a long
time (Kozlowski 1984). In some species tolerant to waterlogging, stomata may reopen after a critical period. In Fraxinus
pennsylvanica Marsh. seedlings, stomata close rapidly following waterlogging but reopen after about 2 weeks (Sena Gomes
and Kozlowski 1980). Stomata of Eucalyptus camaldulensis
Dehnh. reopen about 5 weeks after waterlogging (van der
Moezel et al. 1989). Tang and Kozlowski (1982) reported that
stomata of Quercus macrocarpa Michx. seedlings reopen
about 30 days after flooding. Reopening of stomata did not
occur in severely waterlogged N. solandri or N. menziesii
seedlings during the experiment.
Nothofagus solandri and N. menziesii differed in sensitivity
WATER STRESS AND WATERLOGGING IN NOTHOFAGUS
to waterlogging. Moderate waterlogging resulted in a reduction of gas exchange properties in N. menziesii, whereas N. solandri seedlings were not affected. However, neither species
was well adapted to severe or prolonged waterlogging. Severe
waterlogging reduced A, gsw and grc in both N. solandri and
N. menziesii, but the patterns of reduction differed between the
two species. Regehr et al. (1975) observed a 50% reduction in
A in waterlogging-tolerant Populus deltoides Bartr. ex Marsh.
during complete inundation of the root system for 28 days. In
our study, A was reduced by approximately 50% in N. solandri
and by more than 65% in N. menziesii after only 8 days of
waterlogging. In moderately waterlogging-tolerant Liquidambar styraciflua L. seedlings, A was reduced by 70% over a
9-day flooding period (Pezeshki and Chambers 1985). Zaerr
(1983) showed that following flooding, A decreased to about
half the control rate within 4 days for Pseudotsga menziesii
(Mirb.) Franco and Picea abies (L.) Karst., but did not change
in Pinus sylvestris L. In N. solandri and N. menziesii, the
reduction in A of seedlings under waterlogged conditions was
associated with decreased gsw and with a marked reduction in
grc, indicating that waterlogging resulted in severe damage to
the photosynthetic apparatus of the plants.
In New Zealand, N. solandri is adapted to a wider range of
soil water conditions than N. menziesii. Nothofagus solandri is
usually either the sole or the dominant species in forests of dry
habitats or on poorly drained soils (Wardle 1970, 1984), and it
codominates with N. menziesii at the wetter end of its ecological range. In areas where the climate is dry and where both
species are present, N. solandri is most likely to occupy the
exposed sites, whereas N. menziesii is more or less restricted
to sheltered sites (Wardle 1984). Our results are consistent with
these observations.
Acknowledgments
The authors thank Karl Schasching for his help in seed collection and
for organizing research facilities. We also thank B. Bullsmith, D.
Clark, D. Conder and P. Fuller for technical assistance, and T. Pearson
for help with preparing graphs. O.J. Sun is grateful for the financial
support provided by T.W. Adams Scholarship, BNZ Scholarship for
Forestry Research, Robert C. Bruce Trust and Young Scientists’ Fund
of the Royal Society of New Zealand, and support of NZ Forest
Research Institute in preparing the manuscript. Comments from K.P.
Hogan on the manuscript are greatly appreciated.
References
Araus, J.L. and K.P. Hogan. 1994. Leaf structure and patterns of
photoinhibition in two neotropical palms in clearings and forest
understory during the dry season. Am. J. Bot. 81:726--738.
Björkman, O., S.B. Powles, D.C. Fork and G. Öquist. 1981. Interaction between high irradiance and water stress on photosynthetic
reactions. Year Book, Carnegie Inst., Washington, D.C. 80:57--59.
Buchan, G.D. and K.S. Grewal. 1990. The power-function models for
the soil moisture characteristics. J. Soil Sci. 41:111--117.
Chaves, M.M. 1991. Effects of water deficits on carbon assimilation.
J. Exp. Bot. 42:1--16.
Comstock, J. and J. Ehleringer. 1984. Photosynthetic response to
slowly decreasing leaf water potential in Encelia frutescens. Oecologia 61:241--248.
637
Farquhar, G.D. and T.D. Sharkey. 1982. Stomatal conductance and
photosynthesis. Annu. Rev. Plant Physiol. 33:317--345.
Hanks, S.L. and D.E. Fairbrothers. 1976. Palynotaxonomic investigation of Fagus L. and Nothofagus BL. Light microscopy, scanning
electron microscopy, and computer analysis. In Botanical Systematics 1. Ed. V.H. Heywood. Academic Press, London, U.K.,
pp 11--141.
Hill, R.S. and J. Read. 1991. A revised infrageneric classification of
Nothofagus (Fagaceae). Bot. J. Linn. Soc. 105:37--72.
Ingestad, T. 1971. A definition of optimum nutrient requirements in
birch seedlings. II. Physiol. Plant. 24:118--125.
Innes, K.P. and D. Kelly. 1992. Water potentials in native woody
vegetation during and after a drought in Canterbury. N.Z. J. Bot.
30:81--94.
Jane, G.T. and T.G.A. Green. 1985. Patterns of stomatal conductance
in six evergreen tree species from a New Zealand cloud forest. Bot.
Gaz. 146:413--420.
Jane, G.T. and T.G.A. Green. 1986. Etiology of forest dieback areas
within the Kaimai Range, North Island, New Zealand. N.Z. J. Bot.
24:513--527.
Janiesch, P. 1991. Ecophysiological adaptations of higher plants in
natural communities to waterlogging. In Ecological Responses to
Environmental Stresses. Eds. J. Rozema and J.A.C. Verkleij. Kluwer Academic Publishers, The Netherlands, pp 50--60.
Kozlowski, T.T. 1984. Plant responses to flooding of soil. BioScience
34:162--167.
McGlone, M.S. 1985. Plant biogeography and the late Cenozoic history of New Zealand. N.Z. J. Bot. 23:723--749.
Mildenhall, D.C. 1980. New Zealand late Cretaceous and Cenozoic
plant biogeography: a contribution. Palaeogeogr. Palaeoclimatol.
Palaeoecol. 31:197--233.
Mooney, H.A., O. Björkman and G.J. Collatz. 1977. Photosynthetic
acclimation to temperature and drought in the desert shrub Larrea
divarcata. Year Book, Carnegie Inst., Washington, D.C. 76:328--335.
Myers, B.J. and J.J. Landsberg. 1989. Drought and seedling growth of
two eucalypt species from contrasting habitats. Tree Physiol.
5:207--218.
Ni, B.-R. and S.G. Pallardy. 1991. Response of gas exchange to water
stress in seedlings of woody angiosperms. Tree Physiol. 8:1--9.
Passioura, J.B. 1982. Water in the soil--plant--atmosphere continuum.
In Physiological Plant Ecology II. Water Relations and Carbon
Assimilation. Encyclopedia of Plant Physiology, New Series, Vol.
12B. Eds. O.L. Lange, P.S. Nobel, C.B. Osmond and H. Ziegler.
Springer-Verlag, Berlin, Heidelberg, pp 5--33.
Pereira, J.S. and T.T. Kozlowski. 1977. Variations among woody
angiosperms in relation to flooding. Physiol. Plant. 41:184--192.
Pezeshki, S.R. and J.L. Chambers. 1985. Stomatal and photosynthesis
response of sweet gum Liquidambar styraciflua to flooding. Can. J.
For. Res. 15:371--375.
Philipson, W.R. and M.N. Philipson. 1988. A classification of the
genus Nothofagus (Fagaceae). Bot. J. Linn. Soc. 98:27--36.
Ranney, T.G., T.H. Whitlow and N.L. Bassuk. 1990. Response of five
temperate deciduous tree species to water stress. Tree Physiol.
6:439--448.
Regehr, D.L., F.A. Bazzaz and W.R. Boggess. 1975. Photosynthesis,
transpiration, and leaf conductance of Populus deltoides in relation
to flooding and drought. Photosynthetica 9:52--61.
Schulze, E.-D. and A.E. Hall. 1982. Stomatal responses, water loss
and CO2 assimilation rates of plants in contrasting environments. In
Physiological Plant Ecology II. Water Relations and Carbon Assimilation. Encyclopedia of Plant Physiology, New Series, Vol.
12B. Eds. O.L. Lange, P.S. Nobel, C.B. Osmond and H. Ziegler.
Springer-Verlag, Berlin, Heidelberg, pp 181--230.
638
SUN, SWEET, WHITEHEAD AND BUCHAN
Seiler, J.R. and B.H. Cazell. 1990. Influence of water stress on the
physiology and growth of red spruce seedlings. Tree Physiol. 6:69-77.
Sena Gomes, A.R. and T.T. Kozlowski. 1980. Growth response and
adaptation of Fraxinus pennsylvania seedlings to flooding. Plant
Physiol. 66:267--271.
Sun, O.J. 1993. Genetics and physiology of Nothofagus solandri var.
cliffortioides and N. menziesii----a comparative study. Ph.D. Thesis.
University of Canterbury, Christchurch, New Zealand, 279 p.
Tang, Z.C. and T.T. Kozlowski. 1982. Some physiological and morphological responses of Quercus macrocarpa seedlings to flooding.
Can. J. For. Res. 12:196--202.
Teskey, R.O., J.A. Fites, L.J. Samuelson and B.C. Bongarten. 1986.
Stomatal and nonstomatal limitations to net photosynthesis in Pinus
taeda L. under different environmental conditions. Tree Physiol.
2:131--142.
Tyree, M.T., J. Alexander and J.-L. Machado. 1992. Loss of hydraulic
conductivity due to drought in intact juveniles of Quercus rubra and
Populus deltoides. Tree Physiol. 10:411--415.
Tyree, M.T. and J.S. Sperry. 1989. Vulnerability of xylem to cavitation
and embolism. Annu. Rev. Plant Physiol. Mol. Biol. 40:19--38.
van der Moezel, P.G., L.E. Watson and D.T. Bell. 1989. Gas exchange
responses of two Eucalyptus species to salinity and waterlogging.
Tree Physiol. 5:251--257.
Wardle, J. 1970. The ecology of Nothofagus solandri. 1. The distribution and relationship with other major forest and shrub species. N.Z.
J. Bot. 8:494--531.
Wardle, J. 1984. The New Zealand beeches. Ecology, utilisation and
management. New Zealand Forest Service, Christchurch, New Zealand, 447 p.
Wardle, P. 1967. Biological flora of New Zealand. 2. Nothofagus
menziesii (Hook. f.) Oerst. (Fagaceae). N.Z. J. Bot. 5:276--302.
Zaerr, J.B. 1983. Short-term flooding and net photosynthesis in seedlings of three conifers. For. Sci. 29:71--78.
Zimmermann, U. and J.A. Milburn. 1982. Transport and storage of
water. In Physiological Plant Ecology II. Water Relations and Carbon Assimilation. Encyclopedia of Plant Physiology, New Series Vol.
12B. Eds. O.L. Lange, P.S. Nobel, C.B. Osmond and H. Ziegler.
Springer-Verlag, Berlin, Heidelberg, pp 135--151.